Dictionary of Meaning
<<Back
Please select a letter:
A |
B |
C |
D |
E |
F |
G |
H |
I |
J |
K |
L |
M |
N |
O |
P |
Q |
R |
S |
T |
U |
V |
W |
X |
Y |
Z |
0-9
Click here for Shopping
Proto-Celtic language
*** Shopping-Tip: Proto-Celtic language
{| border="1" cellspacing="0" align="right" cellpadding="2"
! align="center" bgcolor=pink | '''
Language families and languages Language classification'''
|-
|
Indo-European languages Indo-European
Celtic languages Celtic
'''Proto-Celtic'''
|}
The '''Proto-Celtic language''', also called '''Common Celtic''', is the putative ancestor of all the known
Celtic languages. Probably spoken around
9th century BC 800 BC, its lexis can be confidently reconstructed on the basis of the
comparative method of
historical linguistics. Proto-Celtic is a direct daughter-language of
Proto-Indo-European language Proto-Indo-European and is widely regarded as the first of the
Indo-European languages to spread in North-Western and Atlantic Europe. The area in which the language seems to have first become distinguishably Proto-Celtic, as opposed to earlier
Centum dialect, corresponds to the
Hallstatt culture, on the western fringes of the
Urnfield. From roughly 800 BC, this culture by influence of "
Thraco-Cimmerian" elements introduced the
Iron Age to Europe. The contemporary
Cimmerians were variously claimed as ancestors of the
Cimbri,
Sugambri and
Cymru, although other etymologies better explain the latter term (see also
British Israelism).
The reconstruction of Proto-Celtic is currently being undertaken. While Continental Celtic presents much substantiation for
phonology, and some for
morphology (linguistics) morphology, recorded material is largely still too scanty to allow a secure reconstruction of
syntax. Although some complete sentences are recorded in
Gaulish language Gaulish and
Celtiberian language Celtiberian, the oldest substantial Celtic
literature is found in
Old Irish language Old Irish, the earliest recorded of the
Insular Celtic languages.
Phonological reconstruction
Consonants
The phonological changes from
Proto-Indo-European language Proto-Indo-European to Proto-Celtic
consonants may be summarised as follows (an asterisk [*] prior to a letter or word designates that the
phoneme or
lexeme is not attested but is a hypothetical, reconstructed form):
{|class="wikitable"
|-
!PIE
!Proto-Celtic
!Example
|-
| *{{PIE|p}}
| *{{PIE|φ}}
| *{{PIE|ph₂tēr}} > *{{PIE|φatīr}} 'father'
|-
| *{{PIE|t}}
| *{{PIE|t}}
| *{{PIE|treyes}} > *{{PIE|trīs}} 'three'
|-
| *{{PIE|k, ḱ}}
| *{{PIE|k}}
| *{{PIE|kan-}} > *{{PIE|kan-}} 'sing'
*{{PIE|ḱm̥tom}} > *{{PIE|kantom}} 'hundred'
|-
| *{{PIE|kÊ·}}
| *{{PIE|kÊ·}}
| *{{PIE|kʷetwr̥}} > *{{PIE|kʷetwar}} 'four'
|-
| *{{PIE|b}}
| *{{PIE|b}}
| *{{PIE|dÊ°ub-no-}} > *{{PIE|dubno-}} 'deep'
|-
| *{{PIE|d}}
| *{{PIE|d}}
| *{{PIE|derk-}} > *{{PIE|derk-}} 'see'
|-
| *{{PIE|g, ǵ}}
| *{{PIE|g}}
| *{{PIE|gli-}} > *{{PIE|gli-}} 'to glue'
*{{PIE|ǵenu-}} > *{{PIE|genu-}} 'jaw'
|-
| *{{PIE|gÊ·}}
| *{{PIE|b}}
| *{{PIE|gÊ·en-}} > *{{PIE|ben-}} 'woman'
|-
| *{{PIE|bÊ°}}
| *{{PIE|b}}
| *{{PIE|bÊ°er-}} > *{{PIE|ber-}} 'carry'
|-
| *{{PIE|dÊ°}}
| *{{PIE|d}}
| *{{PIE|dʰeh�-}} > *{{PIE|dī-}} 'suck'
|-
| *{{PIE|gʰ, ǵʰ}}
| *{{PIE|g}}
| *{{PIE|gÊ°abÊ°-}} > *{{PIE|gab-}} 'take'
*{{PIE|ǵʰelH-ro-}} > *{{PIE|galaro-}} 'sickness'
|-
| *{{PIE|gÊ·Ê°}}
| *{{PIE|gÊ·}}
| *{{IPA|gʷʰn̥-}} > *{{PIE|gʷan-}} 'kill, wound'
|-
| *{{PIE|s}}
| *{{PIE|s}}
| *{{PIE|seno-}} > *{{PIE|seno-}} 'old'
|-
| *{{PIE|m}}
| *{{PIE|m}}
| *{{PIE|mehâ‚‚tÄ“r}} > *{{PIE|mÄ?tÄ«r}} 'mother'
|-
| *{{PIE|n}}
| *{{PIE|n}}
| *{{PIE|nepÅ?t-}} > *{{PIE|neφūt-}} 'nephew'
|-
| *{{PIE|l}}
| *{{PIE|l}}
| *{{PIE|ligÊ°-}} > *{{PIE|lig-}} 'lick'
|-
| *{{PIE|r}}
| *{{PIE|r}}
| *{{PIE|rēǵ-s}} > *{{PIE|rīgs}} 'king'
|-
| *{{PIE|y}}
| *{{PIE|y}}
| *{{PIE|yuwn̥ko-}} > *{{PIE|yuwanko-}} 'young'
|-
| *{{PIE|w}}
| *{{PIE|w}}
| *{{PIE|wlati-}} > *{{PIE|wlati-}} 'dominion'
|}
In contrast to the parent language, Proto-Celtic does not use
aspiration as a
feature (linguistics) feature for distinguishing
phonemes. So where the Proto-Indo-European
voiced consonant voiced aspirated stops *{{PIE|bʰ}}, *{{PIE|dʰ}}, *{{PIE|gʰ/ǵʰ}} merged with *{{PIE|b}}, *{{PIE|d}}, *{{PIE|g/ǵ}}. The voiced aspirate
labiovelar consonant labiovelar *{{PIE|gʷʰ}} did not merge with *{{PIE|gʷ}}, though: plain *{{PIE|gʷ}} became *{{PIE|b}} in Proto-Celtic, while aspirated *{{PIE|gʷʰ}} became *{{PIE|gʷ}}. Thus, while PIE *{{PIE|gʷen-}} 'woman' became Old Irish ''ben'' and Welsh ''benyw'', PIE *{{PIE|gʷʰn̥-}} 'to kill, to wound' is the source of Old Irish ''gonaid'' and Welsh ''gwanu''.
Proto-Indo-European *{{PIE|p}} was lost in Proto-Celtic, apparently going through the stages *{{PIE|φ}} (as in the table above) and *{{PIE|h}} (perhaps attested by the
toponym Hercynia if this is of Celtic origin) before being lost completely word-initially and between vowels. Adjacent to consonants, Proto-Celtic *{{PIE.html">consonant cluster
clusters *{{PIE|φs}} and *{{PIE|φt}} became *{{PIE|xs}} and *{{PIE|xt}} respectively already in Proto-Celtic. PIE *{{PIE|sp-}} became Old Irish ''s'' and Brythonic ''f''; while Schrijver (1995, 348) argues there was an intermediate stage *{{PIE|sφ-}} (in which *{{PIE|φ}} remained an independent phoneme until after Proto-Insular Celtic had diverged into Goidelic and Brythonic), McCone (1996, 44–45) finds it more economical to believe that *{{PIE|sp-}} remained unchanged in PC, that is, the change *{{PIE|p}} to *{{PIE|φ}} did not happen when *{{PIE|s}} preceded. (Similarly,
Grimm's law did not apply to ''*p, t, k'' after *''s'' in
Proto-Germanic language Germanic.)
{|class="wikitable"
! Proto-Celtic
! Old Irish
! Welsh
|-
| *{{PIE|laφs-}} > *{{PIE|laxs-}} 'shine'
| ''las-aid''
| ''llach-ar''
|-
| *{{PIE|seφtam}} > *{{PIE|sextam}} 'seven'
| ''secht''
| ''seith''
|-
| *{{PIE|sφeret-}} or *{{PIE|speret-}} 'heel'
| ''seir''
| ''ffêr''
|}
In
Gaulish language Gaulish and the
Brythonic languages, a new *{{PIE.html">Welsh language
Welsh ''pedwar'' "four", compared to
Old Irish language Old Irish *''cethair'' and
Latin ''quattuor''. In so far as this new /p/ fills the space in the phoneme inventory which was lost by the disappearence of the equivalent stop in PIE, we may think of this as a
chain shift.
The terms '''P-Celtic''' and '''Q-Celtic''' are useful when we wish to group the Celtic languages according to the way they handle this one phoneme. However a simple division into P- and Q-Celtic may be untenable, as it does not do justice to the evidence of the ancient
Continental Celtic languages. The large number of unusual shared innovations among the
Insular Celtic languages are often also presented as evidence against a P-Celtic ''vs'' Q-Celtic division, but they may instead reflect a common
substratum influence from the pre-Celtic languages of the British Isles [http://www.sciencemag.org/cgi/content/full/288/5469/1158], in which case they would be irrelevant to Celtic language classification.
Q-Celtic languages may also have /p/ in loan words, though in some early borrowings from Welsh into Irish /k/ was used by sound substitution, as in Gaelic ''Cothrige'', an early form of
Saint Patrick "Padraig". Gaelic ''póg'' "kiss" was a later borrowing (from the second word of the Latin phrase ''osculum pacis'' "kiss of peace") at a stage where ''p'' was borrowed directly as ''p'', without substituting ''c''.
Vowels
The Proto-Celtic vowel system is highly comparable to that reconstructed for
Proto-Indo-European language Proto-Indo-European by
Antoine Meillet. Dissimilarities include the incidence of Celtic *''Ä«'' for Proto-Indo-European *''Ä“'' (e.g., Gaulish ''rix'' and Irish ''rÃ'', "king"; compare Latin ''rÄ“x'') and *''Ä?'' in place of *''Å?''.
{|class="wikitable"
! PIE
! Proto-Celtic
! Example
|-
| *{{PIE|a, hâ‚‚e}}
| *{{PIE|a}}
| *{{PIE|hâ‚‚ebon-}} > *{{PIE|abon-}} 'river'
|-
| *{{PIE|Ä?, *ehâ‚‚}}
| *{{PIE|Ä?}}
| *{{PIE|bÊ°rehâ‚‚tÄ“r}} > *{{PIE|brÄ?tÄ«r}} 'brother'
|-
| *{{PIE|e, hâ‚?e}}
| *{{PIE|e}}
| *{{PIE|seno-}} > *{{PIE|seno-}} 'old'
|-
| *{{PIE|"É™"}} (any laryngeal {{PIE|H}} between consonants)
| *{{PIE|a}}
| *{{PIE|ph₂tēr}} > *{{PIE|φatīr}} 'father'
|-
| *{{PIE|Ä“, ehâ‚?}}
| *{{PIE|Ä«}}
| *{{PIE|wērh�o-}} > *{{PIE|wīro-}} 'true'
|-
| *{{PIE|o, Ho, h₃e}}
| *{{PIE|o}}
| *{{PIE|rotos}} > *{{PIE|rotos}} 'wheel'
|-
| rowspan="2" | *{{PIE|Å?, eh₃}}
| in final syllable, *{{PIE|Å«}}
| *{{PIE|nepÅ?t-}} > *{{PIE|neφūt-}} 'nephew'
|-
| elsewhere, *{{PIE|Ä?}}
| *{{PIE|deh₃no-}} > *{{PIE|dÄ?no-}} 'gift'
|-
| *{{PIE|i}}
| *{{PIE|i}}
| *{{PIE|gÊ·itu-}} > *{{PIE|bitu-}} 'world'
|-
| *{{PIE|Ä«, iH}}
| *{{PIE|Ä«}}
| *{{PIE|rÄ«mehâ‚‚}} > *{{PIE|rÄ«mÄ?}} 'number'
|-
| *{{PIE|ai, hâ‚‚ei, ehâ‚‚i}}
| *{{PIE|ai}}
| *{{PIE|kaikos}} > *{{PIE|kaikos}} 'blind'
*{{PIE|sehâ‚‚itlo-}} > *{{PIE|saitlo-}} 'age'
|-
| *{{PIE|(hâ‚?)ei, Ä“i, ehâ‚?i}}
| *{{PIE|Ä“}}
| *{{PIE|deiwos}} > *{{PIE|dēwos}} 'god'
|-
| *{{PIE|oi, Å?i, h₃ei, eh₃i}}
| *{{PIE|oi}}
| *{{PIE|oinos}} > *{{PIE|oinos}} 'one'
|-
| rowspan=2 | *{{PIE|u}}
| before {{PIE|wa}}, {{PIE|o}}
| *{{PIE|yuwn̥kos}} > *{{PIE|yowankos}} 'young'
|-
| elsewhere, *{{PIE|u}}
| *{{PIE|srutos}} > *{{PIE|srutos}} 'stream'
|-
| *{{PIE|Å«, uH}}
| *{{PIE|Å«}}
| *{{PIE|ruHnehâ‚‚}} > *{{PIE|rÅ«nÄ?}} 'mystery'
|-
| *{{PIE|au, hâ‚‚eu, ehâ‚‚u}}
| *{{PIE|au}}
| *{{PIE|tausos}} > *{{PIE|tausos}} 'silent'
|-
| *{{PIE|(hâ‚?)eu, Ä“u, ehâ‚?u}};
*{{PIE|ou, Å?u, h₃eu, eh₃u}}
| *{{PIE|ou}}
| *{{PIE|teutehâ‚‚}} > *{{PIE|toutÄ?}} 'people'
*{{PIE|gÊ·Å?u-}} > *{{PIE|bou-}}
|-
| rowspan=2 | *{{PIE|l̥}}
| before stops, *{{PIE|li}}
| *{{PIE|pl̥th₂nos}} > *{{PIE|φlitanos}} 'wide'
|-
| before other consonants, *{{PIE|al}}
| *{{PIE|klÌ¥yÄ?kos}} > *{{PIE|kalyÄ?kos}} 'rooster'
|-
| rowspan=2 | {{PIE|*r̥}}
| before stops, *{{PIE|ri}}
| *{{PIE|bʰr̥ti-}} > *{{PIE|briti-}} 'act of bearing; mind'
|-
| before other consonants, *{{PIE|ar}}
| *{{PIE|mr̥wos}} > *{{PIE|marwos}} 'dead'
|-
| *{{PIE|m̥}}
| *{{PIE|am}}
| *{{PIE|dm̥-na-}} > *{{PIE|damna-}} 'subdue'
|-
| *{{PIE|n̥}}
| *{{PIE|an}}
| *{{PIE|dn̥t-}} > *{{PIE|dant-}} 'tooth'
|-
| rowspan=2 | *{{PIE|l̥H}}
| before
obstruent consonant obstruents, *{{PIE|la}}
| *{{PIE|wl̥Hti-}} > *{{PIE|wlati-}} 'lordship'
|-
| before
sonorant consonant sonorants, *{{PIE|lÄ?}}
| *{{PIE|plÌ¥Hmehâ‚‚}} > *{{PIE|φlÄ?mÄ?}} 'hand'
|-
| rowspan=2 | *{{PIE|r̥H}}
| before obstruents, *{{PIE|ra}}
| *{{PIE|mr̥Htom}} > *{{PIE|mratom}} 'betrayal'
|-
| before sonorants, *{{PIE|rÄ?}}
| *{{PIE|ǵrÌ¥Hnom}} > *{{PIE|*grÄ?nom}} 'grain'
|-
| *{{PIE|m̥H}}
| *{{PIE|am/mÄ?}}
(presumably same distribution as above)
| (none?)
|-
| *{{PIE|n̥H}}
| *{{PIE|an}} or *{{PIE|nÄ?}}
(presumably same distribution as above)
| probably *{{PIE|gn̥h₃to-}} > *{{PIE|gnato-}} 'knowing'
|}
The vowel *{{PIE|"É™"}} is the so-called "
schwa#Schwa Indogermanicum schwa indogermanicum", now interpreted as a laryngeal between two consonants.
Transition to Welsh
The regular
consonantal
sound changes from Proto-Celtic to the
Welsh language may be summarised in the following table:
{|class="wikitable"
|-
!colspan="1"|Proto-Celtic
consonant
!colspan="1"|
Welsh language Welsh consonant
|-
| *b-
| b
|-
| *-bb-
| b
|-
| *-(V)b(V)-
| f /v/
|-
| *d-
| d
|-
| *-dd-
|
|-
| *-(V)d(V)-
| dd /ð/
|-
| *g-
| g
|-
| *-gg-
| g
|-
| *-(V)g(V)-
| (lost)
|-
| *gw-
| chw /xw/, gw, w
|-
| *h-
| (lost)
|-
| *-h-
| (lost)
|-
| *j-
| i
|-
| *k-
| c
|-
| *-kk-
| g
|-
| *-(V)k(V)-
| ch, h, g
|-
| *kw-
| p
|-
| *-kw-
| b
|-
| *l-
| ll /ɬ/
|-
| *-ll-
| l
|-
| *-(V)l(V)-
| l
|-
| *m-
| m
|-
| *-mb-
| m
|-
| *(C)m
| m
|-
| *-m-
| f
|-
| *n-
| n
|-
| *-n-
| n
|-
| *-nd-
| n
|-
| *-nt-
| nt, nth, nd
|-
| *r-
| rh /r̥/
|-
| *-r-
| r
|-
| *s-
| s, h
|-
| *-s-
| s
|-
| *sk
| sg
|-
| *t
| t
|-
| *-t-
| d, th /θ/
|-
| *w-
| chw /xw/, gw, w
|}
Morphology
The
morphology (linguistics) morphology (structure) of nouns and adjectives demonstrates no arresting alterations from the parent language. The Insular Celtic verb, on the other hand, shows a peculiar feature unknown in any other attested
Indo-European language: verbs have different
Grammatical conjugation conjugational forms depending on whether they appear in absolute initial position in the sentence (Insular Celtic having
Verb Subject Object or VSO word order) or whether they are preceded by a preverbal
grammatical particle particle. The situation is most robustly attested in
Old Irish language Old Irish, but it has remained to some extent in
Scottish Gaelic language Scottish Gaelic and traces of it are present in Middle
Welsh language Welsh as well.
Forms that appear in sentence-initial position are called '''absolute''', those that appear after a particle are called '''conjunct'''. The
paradigm of the
present tense present active voice active indicative mood indicative of the Old Irish verb ''beirid'' "carry" is as follows; the conjunct forms are illustrated with the particle ''nÃ'' "not".
{|class="wikitable"
!
! Absolute
! Conjunct
|-
| 1st person singular || biru "I carry" || nà biur "I do not carry"
|-
| 2nd person singular || biri "you carry" || nà bir "you do not carry"
|-
| 3rd person singular || beirid "s/he carries" || nà beir "s/he does not carry"
|-
| 1st person plural || bermai "we carry" || nà beram "we do not carry"
|-
| 2nd person plural || beirthe "you carry" || nà beirid "you do not carry"
|-
| 3rd person plural || berait "they carry" || nà berat "they do not carry"
|}
In Scottish Gaelic this distinction is still found in the
future tense:
{|class="wikitable"
! Absolute
! Conjunct
|-
| cuiridh "will put" || cha chuir "will not put"
|-
| òlaidh "will drink" || chan òl "will not drink"
|-
| ceannaichidh "will buy" || cha cheannaich "will not buy"
|}
In Middle Welsh, the distinction is seen most clearly in
proverbs following the formula "X happens, Y does not happen" (Evans 1964: 119):
*'''Pereid''' y rycheu, '''ny phara''' a'e goreu "The furrows last, he who made them lasts not"
*'''Trenghit''' golut, '''ny threingk''' molut "Wealth perishes, fame perishes not"
*'''Tyuit''' maban, '''ny thyf''' y gadachan "An infant grows, his swaddling-clothes grow not"
*'''Chwaryit''' mab noeth, '''ny chware''' mab newynawc "A naked boy plays, a hungry boy plays not"
The older analysis of the distinction, as reported by Thurneysen (1946, 360 ff.), held that the absolute endings derive from
Proto-Indo-European language Proto-Indo-European "primary endings" (used in present and future tenses) while the conjunct endings derive from the "secondary endings" (used in past tenses). Thus Old Irish absolute ''beirid'' "s/he carries" was thought to be from *{{PIE|bÊ°ereti}} (compare
Sanskrit language Sanskrit ''bharati'' "s/he carries"), while conjunct ''beir'' was thought to be from *{{PIE|bÊ°eret}} (compare Sanskrit ''a-bharat'' "s/he was carrying").
Today, however, most Celticists agree that Cowgill (1975), following an idea present already in Pedersen (1913, 340 ff.), found the correct solution to the origin of the absolute/conjunct distinction: an
enclitic particle, reconstructed as *{{PIE|es}} after consonants and *{{PIE|s}} after vowels, came in second position in the sentence. If the first word in the sentence was another particle, *{{PIE|(e)s}} came after that and thus before the verb, but if the verb was the first word in the sentence, *{{PIE|(e)s}} was cliticized to it. Under this theory, then, Old Irish absolute ''beirid'' comes from Proto-Celtic *{{PIE|bereti-s}}, while conjunct ''nà beir'' comes from *{{PIE|nī-s bereti}}.
The identity of the *{{PIE|(e)s}} particle remains uncertain. Cowgill suggests it might be a semantically degraded form of *{{PIE|esti}} "is", while Schrijver (1994) has argued it is derived from the particle *{{PIE|eti}} "and then", which is attested in Gaulish.
Continental Celtic languages cannot be shown to have any absolute/conjunct distinction. However, they seem to show only
Subject Verb Object SVO and
Subject Object Verb SOV word orders, as in other Indo-European languages. The absolute/conjunct distinction may thus be an artifact of the VSO word order that arose in Insular Celtic.
References
*Cowgill, Warren (1975). The origins of the Insular Celtic conjunct and absolute verbal endings. In ''Flexion und Wortbildung: Akten der V. Fachtagung der Indogermanischen Gesellschaft, Regensburg, 9.–14. September 1973'', ed. H. Rix, 40–70.
Wiesbaden: Reichert.
*{{cite book | author=Evans, D. Simon | title=A Grammar of Middle Welsh | publisher=
Dublin: Dublin Institute for Advanced Studies | year=1964 | id= }}
*{{cite book | author=McCone, Kim | title=Towards a Relative Chronology of Ancient and Medieval Celtic Sound Change | publisher=
Maynooth: Department of Old and Middle Irish,
St Patrick's College, Maynooth St. Patrick's College | year=1996 | id=ISBN 0-901519-40-5}}
*{{cite book | author=Pedersen, Holger | title=Vergleichende Grammatik der keltischen Sprachen | publisher=2. Band: Bedeutungslehre (Wortlehre).
Göttingen: Vandenhoeck & Ruprecht | year=1913 | id=ISBN 3-525-26119-5}}
*{{cite journal | author=Schrijver, Peter | title=The Celtic adverbs for 'against' and 'with' and the early apocope of *''-i'' | journal=Ériu | year=1994 | volume=45 | pages=151–89}}
*{{cite book | author=Schrijver, Peter | title=Studies in British Celtic Historical Phonology | publisher=
Amsterdam: Rodopi | year=1995 | id=ISBN 90-5183-820-4}}
*{{cite book | author=Thurneysen, Rudolf | title=A Grammar of Old Irish | publisher=Tr. D. A. Binchy and Osborn Bergin. Dublin: Dublin Institute for Advanced Studies | year=1946 | id= }}
See also
*
Italo-Celtic
External links
A reference for Proto-Celtic
vocabulary is provided by the
University of Wales at the following sites:
-
Proto-Celtic English dictionary
-
English Proto-Celtic dictionary
Alternatively, the University of
Leiden provides a Proto-Celtic dictionary:
-
Database query to An etymological lexicon of Proto-Celtic
Category:Celtic languages
Category:Proto-languages
zh:原始凱爾特語
*** Shopping-Tip: Proto-Celtic language
